Impact of a machine learning algorithm on time to palliative care in a primary care population: protocol for a stepped-wedge pragmatic randomized trial.

BACKGROUND: As primary care populations age, timely identification of palliative care need is becoming increasingly relevant. Previous studies have targeted particular patient populations with life-limiting disease, but few have focused on patients in a primary care setting. Toward this end, we propose a stepped-wedge pragmatic randomized trial whereby a machine learning algorithm identifies patients empaneled to primary care units at Mayo Clinic (Rochester, Minnesota, United States) with high likelihood of palliative care need. METHODS: 42 care team units in 9 clusters were randomized to 7 wedges, each lasting 42 days. For care teams in treatment wedges, palliative care specialists review identified patients, making recommendations to primary care providers when appropriate. Care teams in control wedges receive palliative care under the standard of care. DISCUSSION: This pragmatic trial therefore integrates machine learning into clinical decision making, instead of simply reporting theoretical predictive performance. Such integration has the possibility to decrease time to palliative care, improving patient quality of life and symptom burden. TRIAL REGISTRATION: Clinicaltrials.gov NCT04604457 , restrospectively registered 10/26/2020. PROTOCOL: v0.5, dated 9/23/2020.

Rates of Asymptomatic COVID-19 Infection and Associated Factors in Olmsted County, Minnesota, in the Prevaccination Era.

Objective: To estimate rates and identify factors associated with asymptomatic COVID-19 in the population of Olmsted County during the prevaccination era. Patients and Methods: We screened first responders (n=191) and Olmsted County employees (n=564) for antibodies to SARS-CoV-2 from November 1, 2020 to February 28, 2021 to estimate seroprevalence and asymptomatic infection. Second, we retrieved all polymerase chain reaction (PCR)-confirmed COVID-19 diagnoses in Olmsted County from March 2020 through January 2021, abstracted symptom information, estimated rates of asymptomatic infection and examined related factors. Results: Twenty (10.5%; 95% CI, 6.9%-15.6%) first responders and 38 (6.7%; 95% CI, 5.0%-9.1%) county employees had positive antibodies; an additional 5 (2.6%) and 10 (1.8%) had prior positive PCR tests per self-report or medical record, but no antibodies detected. Of persons with symptom information, 4 of 20 (20%; 95% CI, 3.0%-37.0%) first responders and 10 of 39 (26%; 95% CI, 12.6%-40.0%) county employees were asymptomatic. Of 6020 positive PCR tests in Olmsted County with symptom information between March 1, 2020, and January 31, 2021, 6% (n=385; 95% CI, 5.8%-7.1%) were asymptomatic. Factors associated with asymptomatic disease included age (0-18 years [odds ratio {OR}, 2.3; 95% CI, 1.7-3.1] and >65 years [OR, 1.40; 95% CI, 1.0-2.0] compared with ages 19-44 years), body mass index (overweight [OR, 0.58; 95% CI, 0.44-0.77] or obese [OR, 0.48; 95% CI, 0.57-0.62] compared with normal or underweight) and tests after November 20, 2020 ([OR, 1.35; 95% CI, 1.13-1.71] compared with prior dates). Conclusion: Asymptomatic rates in Olmsted County before COVID-19 vaccine rollout ranged from 6% to 25%, and younger age, normal weight, and later tests dates were associated with asymptomatic infection.

Quality control recommendations for RNASeq using FFPE samples based on pre-sequencing lab metrics and post-sequencing bioinformatics metrics.

BACKGROUND: Formalin-fixed, paraffin-embedded (FFPE) tissues have many advantages for identification of risk biomarkers, including wide availability and potential for extended follow-up endpoints. However, RNA derived from archival FFPE samples has limited quality. Here we identified parameters that determine which FFPE samples have the potential for successful RNA extraction, library preparation, and generation of usable RNAseq data. METHODS: We optimized library preparation protocols designed for use with FFPE samples using seven FFPE and Fresh Frozen replicate pairs, and tested optimized protocols using a study set of 130 FFPE biopsies from women with benign breast disease. Metrics from RNA extraction and preparation procedures were collected and compared with bioinformatics sequencing summary statistics. Finally, a decision tree model was built to learn the relationship between pre-sequencing lab metrics and qc pass/fail status as determined by bioinformatics metrics. RESULTS: Samples that failed bioinformatics qc tended to have low median sample-wise correlation within the cohort (Spearman correlation < 0.75), low number of reads mapped to gene regions (< 25 million), or low number of detectable genes (11,400 # of detected genes with TPM > 4). The median RNA concentration and pre-capture library Qubit values for qc failed samples were 18.9 ng/ul and 2.08 ng/ul respectively, which were significantly lower than those of qc pass samples (40.8 ng/ul and 5.82 ng/ul). We built a decision tree model based on input RNA concentration, input library qubit values, and achieved an F score of 0.848 in predicting QC status (pass/fail) of FFPE samples. CONCLUSIONS: We provide a bioinformatics quality control recommendation for FFPE samples from breast tissue by evaluating bioinformatic and sample metrics. Our results suggest a minimum concentration of 25 ng/ul FFPE-extracted RNA for library preparation and 1.7 ng/ul pre-capture library output to achieve adequate RNA-seq data for downstream bioinformatics analysis.

Somatic mutations in benign breast disease tissues and association with breast cancer risk.

BACKGROUND: Benign breast disease (BBD) is a risk factor for breast cancer (BC); however, little is known about the genetic alterations present at the time of BBD diagnosis and how these relate to risk of incident BC. METHODS: A subset of a long-term BBD cohort was selected to examine DNA variation across three BBD groups (42 future estrogen receptor-positive (ER+) BC, 36 future estrogen receptor-negative (ER-) BC, and 42 controls cancer-free for at least 16 years post-BBD). DNA extracted from archival formalin fixed, paraffin-embedded (FFPE) tissue blocks was analyzed for presence of DNA alterations using a targeted panel of 93 BC-associated genes. To address artifacts frequently observed in FFPE tissues (e.g., C>T changes), we applied three filtering strategies based on alternative allele frequencies and nucleotide substitution context. Gene-level associations were performed using two types of burden tests and adjusted for clinical and technical covariates. RESULTS: After filtering, the variant frequency of SNPs in our sample was highly consistent with population allele frequencies reported in 1 KG/ExAC (0.986, p < 1e-16). The top ten genes found to be nominally associated with later cancer status by four of 12 association methods(p < 0.05) were MED12, MSH2, BRIP1, PMS1, GATA3, MUC16, FAM175A, EXT2, MLH1 and TGFB1, although these were not statistically significant in permutation testing. However, all 10 gene-level associations had OR < 1 with lower mutation burden in controls compared to cases, which was marginally statistically significant in permutation testing (p = 0.04). Comparing between the three case groups, BBD ER+ cases were closer to controls in mutation profile, while BBD ER- cases were distinct. Notably, the variant burden was significantly higher in controls than in either ER+ or ER- cases. CD45 expression was associated with mutational burden (p < 0.001). CONCLUSIONS: Somatic mutations were more frequent in benign breast tissue from women who did not develop cancer, opening questions of clonal diversity or immune-mediated restraint on future cancer development. CD45 expression was positively associated with mutational burden, most strongly in controls. Further studies in both normal and premalignant tissues are needed to better understand the role of somatic gene mutations and their contribution to future cancer development.

Quantifying the Importance of COVID-19 Vaccination to Our Future Outlook.

Predictive models have played a critical role in local, national, and international response to the COVID-19 pandemic. In the United States, health care systems and governmental agencies have relied on several models, such as the Institute for Health Metrics and Evaluation, Youyang Gu (YYG), Massachusetts Institute of Technology, and Centers for Disease Control and Prevention ensemble, to predict short- and long-term trends in disease activity. The Mayo Clinic Bayesian SIR model, recently made publicly available, has informed Mayo Clinic practice leadership at all sites across the United States and has been shared with Minnesota governmental leadership to help inform critical decisions during the past year. One key to the accuracy of the Mayo Clinic model is its ability to adapt to the constantly changing dynamics of the pandemic and uncertainties of human behavior, such as changes in the rate of contact among the population over time and by geographic location and now new virus variants. The Mayo Clinic model can also be used to forecast COVID-19 trends in different hypothetical worlds in which no vaccine is available, vaccinations are no longer being accepted from this point forward, and 75% of the population is already vaccinated. Surveys indicate that half of American adults are hesitant to receive a COVID-19 vaccine, and lack of understanding of the benefits of vaccination is an important barrier to use. The focus of this paper is to illustrate the stark contrast between these 3 scenarios and to demonstrate, mathematically, the benefit of high vaccine uptake on the future course of the pandemic.

Statistical analysis of comparative tumor growth repeated measures experiments in the ovarian cancer patient derived xenograft (PDX) setting.

Repeated measures studies are frequently performed in patient-derived xenograft (PDX) models to evaluate drug activity or compare effectiveness of cancer treatment regimens. Linear mixed effects regression models were used to perform statistical modeling of tumor growth data. Biologically plausible structures for the covariation between repeated tumor burden measurements are explained. Graphical, tabular, and information criteria tools useful for choosing the mean model functional form and covariation structure are demonstrated in a Case Study of five PDX models comparing cancer treatments. Power calculations were performed via simulation. Linear mixed effects regression models applied to the natural log scale were shown to describe the observed data well. A straight growth function fit well for two PDX models. Three PDX models required quadratic or cubic polynomial (time squared or cubed) terms to describe delayed tumor regression or initial tumor growth followed by regression. Spatial(power), spatial(power) + RE, and RE covariance structures were found to be reasonable. Statistical power is shown as a function of sample size for different levels of variation. Linear mixed effects regression models provide a unified and flexible framework for analysis of PDX repeated measures data, use all available data, and allow estimation of tumor doubling time.

The DNA Cytosine Deaminase APOBEC3B is a Molecular Determinant of Platinum Responsiveness in Clear Cell Ovarian Cancer.

PURPOSE: Clear cell ovarian carcinoma (CCOC) is an aggressive disease that often demonstrates resistance to standard chemotherapies. Approximately 25% of patients with CCOC show a strong APOBEC mutation signature. Here, we determine which APOBEC3 enzymes are expressed in CCOC, establish clinical correlates, and identify a new biomarker for detection and intervention. EXPERIMENTAL DESIGNS: APOBEC3 expression was analyzed by IHC and qRT-PCR in a pilot set of CCOC specimens (n = 9 tumors). The IHC analysis of APOBEC3B was extended to a larger cohort to identify clinical correlates (n = 48). Dose-response experiments with platinum-based drugs in CCOC cell lines and carboplatin treatment of patient-derived xenografts (PDXs) were done to address mechanistic linkages. RESULTS: One DNA deaminase, APOBEC3B, is overexpressed in a formidable subset of CCOC tumors and is low or absent in normal ovarian and fallopian tube epithelial tissues. High APOBEC3B expression associates with improved progression-free survival (P = 0.026) and moderately with overall survival (P = 0.057). Cell-based studies link APOBEC3B activity and subsequent uracil processing to sensitivity to cisplatin and carboplatin. PDX studies extend this mechanistic relationship to CCOC tissues. CONCLUSIONS: These studies demonstrate that APOBEC3B is overexpressed in a subset of CCOC and, contrary to initial expectations, associated with improved (not worse) clinical outcomes. A likely molecular explanation is that APOBEC3B-induced DNA damage sensitizes cells to additional genotoxic stress by cisplatin. Thus, APOBEC3B is a molecular determinant and a candidate predictive biomarker of the therapeutic response to platinum-based chemotherapy. These findings may have broader translational relevance, as APOBEC3B is overexpressed in many different cancer types.

Poly(adenosine diphosphate ribose) polymerase inhibitors induce autophagy-mediated drug resistance in ovarian cancer cells, xenografts, and patient-derived xenograft models.

BACKGROUND: Poly(adenosine diphosphate ribose) polymerase (PARP) inhibitors exhibit promising activity against ovarian cancers, but their efficacy can be limited by acquired drug resistance. This study explores the role of autophagy in regulating the sensitivity of ovarian cancer cells to PARP inhibitors. METHODS: Induction of autophagy was detected by punctate LC3 fluorescence staining, LC3I to LC3II conversion on Western blot analysis, and electron microscopy. Enhanced growth inhibition and apoptosis were observed when PARP inhibitors were used with hydroxychloroquine, chloroquine (CQ), or LYS05 to block the hydrolysis of proteins and lipids in autophagosomes or with small interfering RNA against ATG5 or ATG7 to prevent the formation of autophagosomes. The preclinical efficacy of the combination of CQ and olaparib was evaluated with a patient-derived xenograft (PDX) and the OVCAR8 human ovarian cancer cell line. RESULTS: Four PARP inhibitors (olaparib, niraparib, rucaparib, and talazoparib) induced autophagy in a panel of ovarian cancer cells. Inhibition of autophagy with CQ enhanced the sensitivity of ovarian cancer cells to PARP inhibitors. In vivo, olaparib and CQ produced additive growth inhibition in OVCAR8 xenografts and a PDX. Olaparib inhibited PARP activity, and this led to increased reactive oxygen species (ROS) and an accumulation of γ-H2AX. Inhibition of autophagy also increased ROS and γ-H2AX and enhanced the effect of olaparib on both entities. Treatment with olaparib increased phosphorylation of ATM and PTEN while decreasing the phosphorylation of AKT and mTOR and inducing autophagy. CONCLUSIONS: PARP inhibitor-induced autophagy provides an adaptive mechanism of resistance to PARP inhibitors in cancer cells with wild-type BRCA, and a combination of PARP inhibitors with CQ or other autophagy inhibitors could improve outcomes for patients with ovarian cancer.

Overcoming platinum resistance in ovarian cancer by targeting pregnancy-associated plasma protein-A.

OBJECTIVES: Inhibition of pregnancy-associated plasma protein-A (PAPP-A), an upstream activator of the insulin-like growth factor (IGF) pathway, is known to augment sensitivity to platinum-based chemotherapy. This study further tests the efficacy of PAPP-A inhibition with a monoclonal antibody inhibitor (mAb-PA) in ovarian cancer (OC) platinum-resistant patient-derived xenograft (PDX) models. METHODS: PAPP-A expression was quantitated in platinum-resistant PDX models by ELISA. A subset with High (n = 5) and Low (n = 2) expression were revived in female SCID/beige mice for studies with either saline, carboplatin/paclitaxel (CP) + mAb-PA, or CP + IgG2a. The primary endpoint was tumor area by ultrasound on day 28 relative to baseline. Conversion to platinum-sensitive was defined by average tumor regression below baseline. Statistical analyses included linear mixed effects modeling and Kaplan Meier curves. Response to therapy was correlated with changes in the ratio of phosphorylated/total AKT and ERK 1/2 using Wes analysis. RESULTS: The addition of mAb-PA to CP induced tumor regression below baseline in one High PAPP-A PDX model; another three models exhibited notable growth inhibition relative to CP + IgG2a. None of the Low PAPP-A PDX models regressed below baseline. The PDX model with the greatest magnitude of tumor regression from baseline after combination therapy was maintained on single agent mAb-PA or IgG2a, but no benefit was observed. Decreased phosphorylation of ERK1/2 correlated with conversion to platinum-sensitive. CONCLUSIONS: The addition of mAb-PA to CP overcame platinum-resistance in one of five High PAPP-A PDX models; three other models demonstrated improved platinum-response. This supports further clinical development of this novel therapeutic.

BRCA1 Deficiency Upregulates NNMT, Which Reprograms Metabolism and Sensitizes Ovarian Cancer Cells to Mitochondrial Metabolic Targeting Agents.

BRCA1 plays a key role in homologous recombination (HR) DNA repair. Accordingly, changes that downregulate BRCA1, including BRCA1 mutations and reduced BRCA1 transcription, due to promoter hypermethylation or loss of the BRCA1 transcriptional regulator CDK12, disrupt HR in multiple cancers. In addition, BRCA1 has also been implicated in the regulation of metabolism. Here, we show that reducing BRCA1 expression, either by CDK12 or BRCA1 depletion, led to metabolic reprogramming of ovarian cancer cells, causing decreased mitochondrial respiration and reduced ATP levels. BRCA1 depletion drove this reprogramming by upregulating nicotinamide N-methyltransferase (NNMT). Notably, the metabolic alterations caused by BRCA1 depletion and NNMT upregulation sensitized ovarian cancer cells to agents that inhibit mitochondrial metabolism (VLX600 and tigecycline) and to agents that inhibit glucose import (WZB117). These observations suggest that inhibition of energy metabolism may be a potential strategy to selectively target BRCA1-deficient high-grade serous ovarian cancer, which is characterized by frequent BRCA1 loss and NNMT overexpression. SIGNIFICANCE: Loss of BRCA1 reprograms metabolism, creating a therapeutically targetable vulnerability in ovarian cancer.

ZC3H18 specifically binds and activates the BRCA1 promoter to facilitate homologous recombination in ovarian cancer.

Reduced BRCA1 expression causes homologous recombination (HR) repair defects in high-grade serous ovarian cancers (HGSOCs). Here, we demonstrate that BRCA1 is transcriptionally activated by a previously unknown function of ZC3H18. We show that ZC3H18 is a DNA-binding protein that interacts with an E2F site in the BRCA1 promoter where it facilitates recruitment of E2F4 to an adjacent E2F site to promote BRCA1 transcription. Consistent with ZC3H18 role in activating BRCA1 expression, ZC3H18 depletion induces BRCA1 promoter methylation, reduces BRCA1 expression, disrupts HR, and sensitizes cells to DNA crosslinkers and poly(ADP-ribose) polymerase inhibitors. Moreover, in patient-derived xenografts and primary HGSOC tumors, ZC3H18 and E2F4 mRNA levels are positively correlated with BRCA1 mRNA levels, further supporting ZC3H18 role in regulating BRCA1. Given that ZC3H18 lies within 16q24.2, a region with frequent copy number loss in HGSOC, these findings suggest that ZC3H18 copy number losses could contribute to HR defects in HGSOC.

Reveal, Don't Conceal: Transforming Data Visualization to Improve Transparency.

Reports highlighting the problems with the standard practice of using bar graphs to show continuous data have prompted many journals to adopt new visualization policies. These policies encourage authors to avoid bar graphs and use graphics that show the data distribution; however, they provide little guidance on how to effectively display data. We conducted a systematic review of studies published in top peripheral vascular disease journals to determine what types of figures are used, and to assess the prevalence of suboptimal data visualization practices. Among papers with data figures, 47.7% of papers used bar graphs to present continuous data. This primer provides a detailed overview of strategies for addressing this issue by (1) outlining strategies for selecting the correct type of figure depending on the study design, sample size, and the type of variable; (2) examining techniques for making effective dot plots, box plots, and violin plots; and (3) illustrating how to avoid sending mixed messages by aligning the figure structure with the study design and statistical analysis. We also present solutions to other common problems identified in the systematic review. Resources include a list of free tools and templates that authors can use to create more informative figures and an online simulator that illustrates why summary statistics are meaningful only when there are enough data to summarize. Last, we consider steps that investigators can take to improve figures in the scientific literature.

Bioinformatics and DNA-extraction strategies to reliably detect genetic variants from FFPE breast tissue samples.

BACKGROUND: Archived formalin fixed paraffin embedded (FFPE) samples are valuable clinical resources to examine clinically relevant morphology features and also to study genetic changes. However, DNA quality and quantity of FFPE samples are often sub-optimal, and resulting NGS-based genetics variant detections are prone to false positives. Evaluations of wet-lab and bioinformatics approaches are needed to optimize variant detection from FFPE samples. RESULTS: As a pilot study, we designed within-subject triplicate samples of DNA derived from paired FFPE and fresh frozen breast tissues to highlight FFPE-specific artifacts. For FFPE samples, we tested two FFPE DNA extraction methods to determine impact of wet-lab procedures on variant calling: QIAGEN QIAamp DNA Mini Kit ("QA"), and QIAGEN GeneRead DNA FFPE Kit ("QGR"). We also used negative-control (NA12891) and positive control samples (Horizon Discovery Reference Standard FFPE). All DNA sample libraries were prepared for NGS according to the QIAseq Human Breast Cancer Targeted DNA Panel protocol and sequenced on the HiSeq 4000. Variant calling and filtering were performed using QIAGEN Gene Globe Data Portal. Detailed variant concordance comparisons and mutational signature analysis were performed to investigate effects of FFPE samples compared to paired fresh frozen samples, along with different DNA extraction methods. In this study, we found that five times or more variants were called with FFPE samples, compared to their paired fresh-frozen tissue samples even after applying molecular barcoding error-correction and default bioinformatics filtering recommended by the vendor. We also found that QGR as an optimized FFPE-DNA extraction approach leads to much fewer discordant variants between paired fresh frozen and FFPE samples. Approximately 92% of the uniquely called FFPE variants were of low allelic frequency range (< 5%), and collectively shared a "C > T|G > A" mutational signature known to be representative of FFPE artifacts resulting from cytosine deamination. Based on control samples and FFPE-frozen replicates, we derived an effective filtering strategy with associated empirical false-discovery estimates. CONCLUSIONS: Through this study, we demonstrated feasibility of calling and filtering genetic variants from FFPE tissue samples using a combined strategy with molecular barcodes, optimized DNA extraction, and bioinformatics methods incorporating genomics context such as mutational signature and variant allelic frequency.

NanoString-based breast cancer risk prediction for women with sclerosing adenosis.

PURPOSE: Sclerosing adenosis (SA), found in » of benign breast disease (BBD) biopsies, is a histological feature characterized by lobulocentric proliferation of acini and stromal fibrosis and confers a two-fold increase in breast cancer risk compared to women in the general population. We evaluated a NanoString-based gene expression assay to model breast cancer risk using RNA derived from formalin-fixed, paraffin-embedded (FFPE) biopsies with SA. METHODS: The study group consisted of 151 women diagnosed with SA between 1967 and 2001 within the Mayo BBD cohort, of which 37 subsequently developed cancer within 10 years (cases) and 114 did not (controls). RNA was isolated from benign breast biopsies, and NanoString-based methods were used to assess expression levels of 61 genes, including 35 identified by previous array-based profiling experiments and 26 from biological insight. Diagonal linear discriminant analysis of these data was used to predict cancer within 10 years. Predictive performance was assessed with receiver operating characteristic area under the curve (ROC-AUC) values estimated from 5-fold cross-validation. RESULTS: Gene expression prediction models achieved cross-validated ROC-AUC estimates ranging from 0.66 to 0.70. Performing univariate associations within each of the five folds consistently identified genes DLK2, EXOC6, KIT, RGS12, and SORBS2 as significant; a model with only these five genes showed cross-validated ROC-AUC of 0.75, which compared favorably to risk prediction using established clinical models (Gail/BCRAT: 0.57; BBD-BC: 0.67). CONCLUSIONS: Our results demonstrate that biomarkers of breast cancer risk can be detected in benign breast tissue years prior to cancer development in women with SA. These markers can be assessed using assay methods optimized for RNA derived from FFPE biopsy tissues which are commonly available.

Postlactational involution biomarkers plasminogen and phospho-STAT3 are linked with active age-related lobular involution.

PURPOSE: Breast terminal duct lobular units undergo two distinctive physiological processes of involution: age-related lobular involution (LI), which is gradual and associated with decreased breast cancer risk, and postlactational involution, which is relatively precipitous, occurs with weaning, and has been associated with potentiation of tumor aggressiveness in animal models. Here we assessed whether markers of postlactational involution are associated with ongoing LI in a retrospective tissue cohort. METHODS: We selected 57 women from the Mayo Clinic Benign Breast Disease Cohort who underwent multiple biopsies and who were average age 48 at initial biopsy. Women were classified as having progressive or non-progressive LI between initial and subsequent biopsy. Serial tissue sections were immunostained for plasminogen, matrix metalloproteinase 9 (MMP-9), phospho-STAT3 (pSTAT3), tenascin C, Ki67, CD44, cytokeratin 14 (CK14), cytokeratin 19 (CK19), and c-myc. All but Ki67 were digitally quantified. Associations between maximal marker expression per sample and progressive versus non-progressive LI were assessed using logistic regression and adjusted for potential confounders. RESULTS: While no biomarker showed statistically significant association with LI progression when evaluated individually, lower expression of pSTAT3 (OR 0.35, 95% CI 0.13-0.82, p = 0.01) and higher expression of plasminogen (OR 2.89, 95% CI 1.14-8.81, p = 0.02) were associated with progressive LI in models simultaneously adjusted for all biomarkers. Sensitivity analyses indicated that the strengthening in association for pSTAT3 and plasminogen with progressive LI was due to collinearity between these two markers. CONCLUSIONS: This is the first study to identify biomarkers of active LI. Our findings that plasminogen and pSTAT3 are significantly associated with LI suggest that they may represent signaling nodes or biomarkers of pathways common to the processes of postlactational involution and LI.

Breast Cancer Risk and Progressive Histology in Serial Benign Biopsies.

Background: More than 1 million women per year in the United States with benign breast biopsies are known to be at elevated risk for breast cancer (BC), with risk stratified on histologic categories of epithelial proliferation. Here we assessed women who had serial benign biopsies over time and how changes in the histologic classification affected BC risk. Methods: In the Mayo Clinic Benign Breast Disease Cohort of 13 466 women, 1414 women had multiple metachronous benign biopsies (10.5%). Both initial and subsequent biopsies were assessed histologically. BC risk for clinical and prognostic factors was assessed using subdistribution models to account for competing risks, and logistic regression/Wilcoxon/chi-square tests to assess covariates. All statistical tests were two-sided. Results: Breast cancer risk for women with serial biopsies, stratified by histologic category in the later biopsies, was similar to women with a single biopsy. We found that changes in histological category between initial and subsequent biopsy statistically significantly impacted BC risk. Women with nonproliferative initial findings and subsequent proliferative findings had an increased risk (hazard ratio [HR] = 1.77, 95% confidence interval [CI] = 1.06 to 2.94, P = .03) compared with no change. Among women with proliferative disease without atypia at initial biopsy, risk decreased if later biopsy regressed to nonproliferative (HR = 0.49, 95% CI = 0.25 to 0.98) and increased if later biopsy showed progression to atypical hyperplasia (HR = 1.49, 95% CI = 0.73 to 3.05) compared with no change ( P = .04). Conclusions: We found that breast cancer risk increases in women with progressive epithelial proliferation over time and decreases in women whose biopsies show less proliferation. This finding has important implications for effective clinical management of the 100 000 women per year who have multiple benign breast biopsies.

Pooled Clustering of High-Grade Serous Ovarian Cancer Gene Expression Leads to Novel Consensus Subtypes Associated with Survival and Surgical Outcomes.

Purpose: Here we assess whether molecular subtyping identifies biological features of tumors that correlate with survival and surgical outcomes of high-grade serous ovarian cancer (HGSOC).Experimental Design: Consensus clustering of pooled mRNA expression data from over 2,000 HGSOC cases was used to define molecular subtypes of HGSOCs. This de novo classification scheme was then applied to 381 Mayo Clinic HGSOC patients with detailed survival and surgical outcome information.Results: Five molecular subtypes of HGSOC were identified. In the pooled dataset, three subtypes were largely concordant with prior studies describing proliferative, mesenchymal, and immunoreactive tumors (concordance > 70%), and the group of tumors previously described as differentiated type was segregated into two new types, one of which (anti-mesenchymal) had downregulation of genes that were typically upregulated in the mesenchymal subtype. Molecular subtypes were significantly associated with overall survival (P < 0.001) and with rate of optimal surgical debulking (≤1 cm, P = 1.9E-4) in the pooled dataset. Among stage III-C or IV Mayo Clinic patients, molecular subtypes were also significantly associated with overall survival (P = 0.001), as well as rate of complete surgical debulking (no residual disease; 16% in mesenchymal tumors compared with >28% in other subtypes; P = 0.02).Conclusions: HGSOC tumors may be categorized into five molecular subtypes that associate with overall survival and the extent of residual disease following debulking surgery. Because mesenchymal tumors may have features that were associated with less favorable surgical outcome, molecular subtyping may have future utility in guiding neoadjuvant treatment decisions for women with HGSOC. Clin Cancer Res; 23(15); 4077-85. ©2017 AACR.

Mammographic breast density and risk of breast cancer in women with atypical hyperplasia: an observational cohort study from the Mayo Clinic Benign Breast Disease (BBD) cohort.

BACKGROUND: Atypical hyperplasia (AH) and mammographic breast density (MBD) are established risk factors for breast cancer (BC), but their joint contributions are not well understood. We examine associations of MBD and BC by histologic impression, including AH, in a subcohort of women from the Mayo Clinic Benign Breast Disease Cohort. METHODS: Women with a diagnosis of BBD and mammogram between 1985 and 2001 were eligible. Histologic impression was assessed via pathology review and coded as non-proliferative disease (NP), proliferative disease without atypia (PDWA) and AH. MBD was assessed clinically using parenchymal pattern (PP) or BI-RADS criteria and categorized as low, moderate or high. Percent density (PD) was also available for a subset of women. BC and clinical information were obtained by questionnaires, medical records and the Mayo Clinic Tumor Registry. Women were followed from date of benign biopsy to BC, death or last contact. Standardized incidence ratios (SIRs) compared the observed number of BCs to expected counts. Cox regression estimated multivariate-adjusted MBD hazard ratios. RESULTS: Of the 6271 women included in the study, 1132 (18.0%) had low MBD, 2921 (46.6%) had moderate MBD, and 2218 (35.4%) had high MBD. A total of 3532 women (56.3%) had NP, 2269 (36.2%) had PDWA and 470 (7.5%) had AH. Over a median follow-up of 14.3 years, 528 BCs were observed. The association of MBD and BC risk differed by histologic impression (p-interaction = 0.03), such that there was a strong MBD and BC association among NP (p < 0.001) but non-significant associations for PDWA (p = 0.27) and AH (p = 0.96). MBD and BC associations for AH women were not significant within subsets defined by type of MBD measure (PP vs. BI-RADS), age at biopsy, number of foci of AH, type of AH (lobular vs. ductal) and body mass index, and after adjustment for potential confounding variables. Women with atypia who also had high PD (>50%) demonstrated marginal evidence of increased BC risk (SIR 4.98), but results were not statistically significant. CONCLUSION: We found no evidence of an association between MBD and subsequent BC risk in women with AH.